Download PDF Download

Focus
Bank and the International Monetary Fund. As a
condition for the loan, the government signed up to a
three-year Memorandum of Economic and Financial
Policies, committing Greece to sweeping spending
cuts, steep tax increases, and an ambitious programme of structural reforms (IMF 2010; EC 2010).
The Greek programme was updated in July 2011,
when the euro area summit conceded lower interest
rates and a longer repayment period (CEU 2011a),
and again in October 2011, when the European summit opened the way to a new loan and a negotiated
reduction in the nominal value of Greek government
bonds, colloquially known as a ‘haircut’ (CEU
2011b).
THE CRISIS AND TAX EVASION
IN GREECE: WHAT ARE THE
DISTRIBUTIONAL IMPLICATIONS?
MANOS MATSAGANIS,
CHRYSA LEVENTI* AND
MARIA FLEVOTOMOU**
Introduction
As every European who occasionally switches on her
TV set must be aware, Greece is in the throes of a dramatic crisis. This started off in 2009 as a fiscal crisis,
soon turned into a sovereign debt crisis, and finally
mutated into a full-blown recession, unprecedented in
depth and duration. At the time of writing (May
2012), the Greek economy had already been in recession for four consecutive years, and showed few signs
of recovery. The latest official figures (Bank of Greece
2012) estimated the size of (negative) growth in 2011
at – 6.9 percent, and bleakly forecast a further – 5.5 percent in 2012. Overall, GDP looks set to
contract by as much as 17.4 percent in real terms in
2012 versus 2008. So deep and drawn out a recession
had simply no precedence in the country’s economic
history at peacetime.
Fiscal consolidation, a crucial part of the programme,
proved moderately successful: revenues rose from
38.0 percent of GDP in 2009 to 41.0 percent in 2011,
while expenditure fell from 53.8 percent of GDP in
2009 to 50.3 percent in 2011 (IMF 2012, 9). Most of
the deficit reduction (about 5 percent of GDP) was
actually achieved in 2010, moving the OECD to
observe that “no other OECD country has achieved
such a fiscal improvement in a single year over the
past three decades” (OECD 2011, 12).
As recognised from the outset, the fight against tax
evasion was to play a crucial role. Firstly, by virtue of
its sheer size: at an estimated 27.5 percent of GDP in
the period 1999–2007, the informal economy in
Greece was larger than any other EU country
(Schneider 2012); the VAT tax gap in 2006 was 30 percent, compared to an EU average of 12 percent; while
the tax debt as a share of annual net tax revenue in
Greece was 72.2 percent in 2011, compared to an
OECD average of 12.3 percent (IMF 2012, 9). In this
context, the scope of improvement was great: the
OECD reckoned that “if Greece collected its VAT,
social security contributions and corporate income
tax with the average efficiency of OECD countries,
tax revenues could rise by nearly 5 percent of GDP”
(OECD 2011, 15).
In May 2010, at the height of the debt crisis, the
Greek government negotiated a 110 billion euro loan
with the European Union, the European Central
* Athens University of Economics and Business, Athens.
** Bank of Greece, Athens.
Our earlier research was supported by the European Commission
and the General Secretariat of Research and Technology of the
Hellenic Republic, and was presented in conferences and seminars held in Buenos Aires (July 2009), London (February 2009),
Milan (June 2008), and Athens (April 2008). We are grateful to
participants for comments and suggestions. We particularly thank
Andrea Brandolini, Nikos Christodoulakis, Carlo Fiorio, Orsolya
Lelkes, Daniela Mantovani, Vassilis Monastiriotis, Emmanuel
Saez, Panos Tsakloglou and Alberto Zanardi. The results presented here are based on an extension of the earlier study,
supported by the European Commission ‘Integrating and
Strengthening the European Research Area’ programme (project
no. 028412), and by the General Secretariat of Research and
Technology of the Hellenic Republic (grant no. 03EΔ319/8.3.1).
We use the European tax-benefit model EUROMOD. The model
is continually being improved and updated; the results presented
here represent the best available at the time of writing. The usual
disclaimer applies.
CESifo Forum 2/2012
Equally obvious is the social and political importance
of progress in the fight against tax evasion. In the
words of the OECD: “a decisive reduction in tax eva-
26
Focus
sion is indispensable for fairness and [the] acceptance
of the broader fiscal consolidation effort. [...]
Reduction in tax evasion has become a major yardstick in measuring the success of the adjustment programme for many observers” (OECD 2011, 15).
es its main findings. The final section reflects on the
policy implications of our findings, the limitations of
our approach, and issues for further research.
Methodology and data
Progress on that front to date has been at best limited. For example, the latest review of the economic
adjustment programme for Greece by the European
Commission concluded that “fiscal consolidation
was held back by a less than successful fight against
tax evasion” (EC 2011, 17). Even more significantly,
this also seems to be the perception of most observers
at home and abroad. Evidence for the latter includes
a recent interview given by Christine Lagarde,
Managing Director of the International Monetary
Fund, to the British daily The Guardian (25 May
2012).1
There is compelling evidence that the rate of underreporting of wages and salaries is much lower than in
the case of self-employment earnings. The analysis of
US tax audit data collected under the Taxpayer
Compliance Measurement Program (TCMP) in 1988
estimated the former at 0.5 percent and the latter at
58.6 percent (Slemrod and Yitzhaki 2002). Similar
data from the successor to TCMP, the National
Research Program (NCP), estimated that 57 percent
of self-employment income was under-reported, compared to 1 percent of wages and salaries (Slemrod
2007). These findings are supported by studies from
other countries, or using different research designs (or
both). Pissarides and Weber (1989) found that the
self-employed in Britain spent a higher share of their
reported income on food (other things such as household characteristics being equal), and attributed this
to income under-reporting, rather than a higher
propensity to consume food – a finding later replicated by Lyssiotou et al. (2004). Feldman and Slemrod
(2007) used this insight to analyse the relationship
between charitable contributions and reported
income, and argued that the higher contributions of
the self-employed at similar levels of reported
incomes could only be explained by higher income
under-reporting. In Italy, Fiorio and D’Amuri (2005)
estimated the rate of under-reporting of self-employment income around the median of the distribution at
27.7 percent, compared to 1.9 percent for income
from wages and salaries, while Marino and Zizza
(2008) found self-employed earnings to be underreported by as much as 56.3 percent. In Hungary,
Benedek and Lelkes (2011) showed that 67 percent of
self-employment income was under-reported, compared to 4 percent of wages and salaries. In Greece,
Matsaganis and Flevotomou (2010) estimated these
rates at 24.4 percent and 0.6 percent respectively.
Our paper focuses on one aspect of tax evasion in
Greece, namely its distributional implications. It
builds on an earlier study (Matsaganis and
Flevotomou 2010), estimating the size and distribution of evasion of personal income tax in Greece in
2004, by extending it to later years drawing on a new,
larger sample of unaudited tax returns filed in
2007–2011 (incomes earned in 2006–2010). The paper
combines an estimation of non-compliance patterns
in terms of income under-reporting, with an estimation of the distribution of gains from tax evasion in
the general population using a tax-benefit model.
We estimate the average rate of under-reporting in
2006 at 11.8 percent, resulting in a shortfall in tax
receipts of 27.8 percent. We show that tax evasion
causes inequality to rise, and the tax system to
become significantly less progressive. We find little
evidence of any significant change in patterns of
income under-reporting since the onset of the current crisis.
The paper is structured as follows: the second section
explains the methodology of the study and presents
the data. Section three reports its results and discuss1
The interview contained the following exchange, widely reported in
Greece:
“Lagarde: Do you know what? As far as Athens is concerned, I
also think about all those people who are trying to evade tax all the
time. All these people in Greece who are trying to evade tax.
Guardian: Even more than you think about all those now struggling to survive without jobs or public services?
Lagarde: I think of them equally. And I think they should also help
themselves collectively.
Guardian: How?
Lagarde: By all paying their tax. Yeah”.
(see “Christine Lagarde: Can the Head of the IMF Save the
Euro?”, The Guardian, 25 May 2012, http://www.guardian.co.uk/
world/2012/may/25/christine-lagarde-imf-euro).
While the evidence on patterns of non-compliance by
income source seems robust, this is not the case with
respect to non-compliance by income class. While the
theory predicts that tax evasion should generally rise
with income (Andreoni et al. 1998), the empirical evidence is mixed. Christian (1994) used data from the
1988 TCMP study to show that, relative to the size of
their true income, higher-income taxpayers evaded less
27
CESifo Forum 2/2012
Focus
under-estimation of the regressive impact of tax evasion. On the contrary, we had direct access to a large
sample of tax returns, provided to us in anonymised
form by the Ministry of Finance.
than those on lower incomes. However, as Slemrod
(2007) has shown, that study classified taxpayers with
high permanent income reporting business losses as
low incomes, while it failed to account for illegal tax
shelters and for non-compliance in partnership and
corporate tax returns. Fiorio and D’Amuri (2005)
found that the share of unreported income in Italy fell
with income. In contrast, Pashardes and Polycarpou
(2008) showed that, once corrected for tax evasion, the
income distribution in Cyprus was less equal than the
distribution of reported incomes. In view of the above,
we focus on under-reporting by income source, assuming no variation by income class (except, of course,
that resulting from composition effect, i.e. the distribution of income by income source).
More specifically, our work draws on two sets of data:
(a) a large sample of unaudited income tax returns
filed in 2007 (incomes earned in 2006) and (b) the
European Union Survey of Income and Living
Conditions (EU-SILC) of 2007 (incomes earned in
2006). The sample of tax returns covers 301,577 tax filers and 96,451 dependent children in 196,742 tax units
(3.6 percent of population), while EU-SILC 2007 contains detailed information on personal incomes and
demographic characteristics of 14,759 individuals in
5,643 households (0.13 percent of population).
Our paper builds on the method applied in Fiorio and
D’Amuri (2005). We also compare data from an
income survey to a sample of tax returns, and assume
that taxpayers concealing part of their income from
tax authorities might consider declaring a higher figure to an anonymous interviewer. Nevertheless, the
fact that Fiorio and D’Amuri (2005) had no direct
access to their sample of tax data forced them to apply
a post-stratification procedure that implicitly assumes
away re-ranking effects, which in turn leads to an
We analyse recent trends in income under-reporting
using the same sample of tax returns filed in 2011
(incomes earned in 2010). Since EU-SILC 2011
(incomes earned in 2010) had not been released at the
time of writing, EU-SILC 2007 income variables were
uprated to 2010 using official estimates provided by
the Hellenic Statistical Authority and the Bank of
Greece (Matsaganis and Leventi 2011). The relevant
tax schedules for 2006 and 2010 are shown in Table 1.
Table 1
Income tax brackets and marginal tax rates
A. Financial year 2007 (incomes earned in 2006)
Income brackets (€ p.a.)
from
to
Tax rate (in %)
0
9,500
0
9,501
13,000
15
13,001
23,000
30
23,001
…
40
B. Financial year 2011 (incomes earned in 2010)
Income brackets (€ p.a.)
from
to
Tax rate (in %)
0
12,000
0
12,001
16,000
18
16,001
22,000
24
22,001
26,000
26
26,001
32,000
32
32,001
40,000
36
40,001
60,000
38
60,001
100,000
40
100,001
…
45
Notes: Personal income tax is individual. Spouses file a joint income tax return, but their income is separately
recorded and individually taxed. The tax unit for the assessment of tax allowances and credits includes spouse
and dependent child(ren). In 2007 the zero-tax threshold was €11,000 for employees or pensioners, and was
raised for taxpayers with dependent children (by €1,000 for one child, by €2,000 for two children, by €10,000
for three children, and by an extra €1,000 for each subsequent child). In 2010 it was €12,000 for all taxpayers,
and was raised for taxpayers with dependent children (by €1,500 for one child, by €3,000 for two children, by
€11,500 for three children, and by an extra €2,000 for each subsequent child).
Source: Own calculations using EUROMOD version F4.32.
CESifo Forum 2/2012
28
Focus
In order to make the two samples comparable, we
restrict the EU-SILC sample to those eligible for submitting a tax return. In view of current tax rules, we
narrowly define tax filers as those meeting at least one
of the following criteria: (a) wage/salary earners with
an annual income above 6,000 euros; (b) farmers
earning more than 3,000 euros per year; and (c) persons with non-zero self-employment income. These
rules are applied both to EU-SILC and the tax returns
sample, reducing their sizes to 7,382 and 219,392 individuals respectively.
Adjustment factors are ratios of income reported in tax
returns to that observed in EU-SILC. More formally, let
i denote region and j income source. Let YTi,j denote
average income observed in region i by source of
income j in EU-SILC, and YRi,j denote the corresponding average income as reported in tax returns.
Adjustment factors are then defined as ai,j = YRi,j /YTi,j .
Since tax returns are cross-checked against the records
of benefit-paying agencies, it is impossible to misreport pension incomes in tax returns, except due to
measurement (e.g. recall) error. In view of this fact, we
have ignored over-reporting of pension incomes
(4.2 percent in 2006). We have also ignored slight rates
of over-reporting for wages and salaries in Athens
and Northern region (1 percent and 1.6 percent
respectively in 2006), setting the relevant factors to
one. The resulting adjustment factors by income
source and region are shown in Table 2.
We ensure that income variables in the two datasets
are consistently defined: in tax returns, incomes are
reported gross of income tax and net of social insurance contributions; in EU-SILC, incomes are reported net of income tax and social insurance contributions. We use the European tax-benefit model
EUROMOD to compute income taxes, which are
then added to net incomes.2
The adjustment factors are used to estimate a ‘synthetic’ income distribution (i.e. adjusted for underreporting) in EU-SILC data (say ЎRi,j, where ЎRi,j = ai,j
× YTi,j). In order to draw out the implications of
income under-reporting for the resulting distribution
of post-tax disposable incomes, and in terms of tax
evaded, we use the European tax-benefit model
EUROMOD.
With respect to correct incomes for tax evasion, we
allocate the reference population into 16 categories
(combinations of 4 macro regions and 4 income
sources). The macro regions are Greater Athens,
Northern, Central and the Islands. The four income
sources are wages and salaries, pensions, farming and
self-employment earnings.
In order to minimise measurement errors, in particular the unreliability of income surveys at the bottom
of the income distribution, we restrict our comparison to employment income and pensions above
6,000 euros per year, and farming and self-employment earnings above 3,000 euros per year.
2
Results and discussion
Table 3 shows how under-reporting varies by income
group. The extent of income under-reporting seems to
be greatest at the two ends of the income distribution.
We (conservatively) estimate the average rate of
See https://www.iser.essex.ac.uk/euromod.
Table 2
Adjustment factors
A. Incomes earned in 2006
Greater Athens
Northern
Central
Islands
Wages/salaries
1.000
1.000
0.884
0.846
Pensions
1.000
1.000
1.000
1.000
Farming
0.533
0.574
0.489
0.490
Self-employment
0.702
0.701
0.624
0.553
B. incomes earned in 2010
Greater Athens
Northern
Central
Islands
Wages/salaries
1.000
1.000
0.886
0.831
Pensions
1.000
1.000
1.000
1.000
Farming
0.513
0.561
0.448
0.482
Self-employment
0.688
0.702
0.617
0.548
Notes: The adjustment factors are multiplied by survey incomes in order to derive a distribution of tax reported
incomes.
Source: Own calculations using EUROMOD version F4.32.
29
CESifo Forum 2/2012
Focus
Table 3
Under-reporting by level of income
A. Incomes earned in 2006
B. Incomes earned in 2010
EU-SILC
Tax data
UnderEU-SILC
Tax data
Under(€)
(€)
reporting (%)
(€)
(€)
reporting (%)
Decile 1
2,128
1,624
23.7
2,277
1,754
22.9
Decile 2
4,049
3,604
11.0
4,416
3,925
11.1
Decile 3
5,884
5,234
11.1
6,402
5,650
11.7
Decile 4
8,050
7,256
9.9
8,862
7,968
10.1
Decile 5
9,886
8,946
9.5
10,689
9,824
8.1
Decile 6
12,141
10,930
10.0
13,080
11,822
9.6
Decile 7
15,217
13,507
11.2
16,389
14,598
10.9
Decile 8
19,635
17,508
10.8
21,011
18,818
10.4
Decile 9
25,843
23,578
8.8
27,649
25,143
9.1
Decile 10
51,988
43,886
15.6
56,795
47,772
15.9
Top 1%
137,087
103,706
24.4
152,517
114,556
24.9
Top 0.1%
315,900
230,253
27.1
352,991
246,652
30.1
Total
15,422
13,600
11.8
16,739
14,723
12.0
Notes: Mean non-equivalised annual personal income by decile in current terms. Income deciles were
constructed excluding those earning zero or negative incomes. Income is adjusted for under-reporting using
the adjustment factors by region and income source as shown in Table 2. All figures are in nominal prices.
Source: Own calculations using EUROMOD version F4.32.
higher under tax evasion than it would have been
under full compliance, the relative poverty line is
also higher (by around 1.5 percent). Nonetheless,
our results suggest that tax evasion causes relative
poverty to rise slightly. All three inequality indicators (Gini, S80/S20, coefficient of variation) have
higher values under tax evasion than under full
compliance, indicating that the former results in a
more unequal income distribution. Finally, our tax
progressivity and redistribution indices (Kakwani,
Reynolds-Smolensky) imply that income underreporting renders the tax system considerably more
regressive.
under-reporting for the entire population at about
12 percent.
Table 4 estimates taxable income and tax receipts
under full compliance and tax evasion respectively.
We show that tax evasion raises average disposable
income by around 4.5 percent; and that it reduces the
tax yield by approximately 28 percent (in 2006, rising
to 30 percent in 2010).
Table 5 presents the distributional implications of
tax evasion in terms of poverty, inequality, and tax
progressivity. Since household disposable income is
Table 4
Income tax variables under full compliance and tax evasion
A. Incomes earned in 2006
B. Incomes earned in 2010
Full
Tax
Full
Tax
compliance
evasion
Shortfall
compliance
evasion
Shortfall
(€)
(€)
(in %)
(€)
(€)
(in %)
Reported income
15,422
13,600
– 11.8
16,739
14,723
– 12.0
Taxable income
15,621
13,701
– 12.3
16,746
14,611
– 12.8
Tax allowances
2,354
2,337
– 0.7
3,205
3,180
– 0.8
Tax reductions
240
238
– 0.9
268
265
– 0.8
Tax due (non-zero)
4,261
3,605
– 15.4
4,263
3,476
– 18.5
Tax due (all)
1,143
825
– 27.8
1,056
736
– 30.3
Disposable income
11,993
12,513
+ 4.3
13,063
13,632
+ 4.4
Notes: Mean non-equivalised annual personal income in current terms. ‘Full compliance’ provides an estimate
of income tax variables assuming incomes are reported to tax authorities as observed in the survey. ‘Tax
evasion’ provides estimates of the same variables assuming incomes are under-reported to tax authorities as
implied by the adjustment factors as shown in Table 2. The share of positive non-zero income earners paying
non-zero tax in 2006 was 40.3% and 34.4% under full compliance and tax evasion respectively (37.2% and
31.9% in 2010). All figures are in nominal prices.
Source: Own calculations using EUROMOD version F4.32.
CESifo Forum 2/2012
30
Focus
Table 5
Fiscal and distributional implications of tax evasion
A. Incomes earned in 2006
B. Incomes earned in 2010
Full
Tax
Full
Tax
compliance
evasion
Difference compliance
evasion
Difference
(€)
(in %)
(€)
(€)
(in %)
(€)
Tax receipts (€ million)
12,131
8,753
– 27.8
11,210
7,817
– 30.3
Poverty line (€ p.a.)
6,041
6,146
+ 1.7
6,600
6,695
+ 1.4
poverty rate
19.7
20.0
+ 1.4
20.3
20.9
+ 2.9
Gini
0.340
0.357
+ 4.9
0.342
0.361
+ 5.4
S80/S20
5.845
6.262
+ 7.1
5.900
6.373
+ 8.0
Coefficient of variation
0.771
0.854
+ 10.8
0.761
0.861
+ 13.3
Kakwani
0.031
0.021
– 31.6
0.030
0.019
– 35.7
Reynolds-Smolensky
0.052
0.035
– 32.2
0.049
0.031
– 36.5
Notes: ‘Full compliance’ provides an estimate of income tax variables assuming incomes are reported to tax
authorities as observed in the survey. ‘Tax evasion’ provides estimates of the same variables assuming
incomes are under-reported to tax authorities as implied by the adjustment factors shown in Table 2. Fiscal
effects (i.e. tax receipts) are non-equivalised. Distributional effects are computed on the basis of equivalised
household disposable incomes. The poverty line is set at 60% of median equivalised household disposable
income, and is calculated separately under full compliance and tax evasion. All figures are in nominal prices.
Source: Own calculations using EUROMOD version F4.32.
These results confirm our earlier findings that farming incomes and self-employment earnings account
for the bulk of tax evasion in Greece, and that the
pattern of income under-reporting by income class is
U-shaped (Matsaganis and Flevotomou 2010).
Moreover, although the crisis does not appear to be
having a massive effect on under-reporting trends,
changes in tax policy (i.e. the personal income tax
schedule) appear to have caused a roughly similar
level of income under-reporting (circa 12 percent) to
translate into significantly higher loss of tax receipts
in 2010 than in 2006 (30.3 percent vs. 27.8 percent
respectively). The finding that the rate of income
under-reporting at the top of the distribution seems
to have risen recently is another pointer in the same
direction.
Yitzhaki 2002; Sandmo 2005). Although the analysis
of dynamic effects lies beyond the scope of this paper,
we need to recognise that the implications of tax evasion exceed those that can be shown with a static arithmetical recalculation of the income distribution.
While our approach focuses on personal income tax,
the distributional impact of evading other taxes (e.g.
company tax, capital tax, value added tax) is likely to
reinforce these effects. Evasion of social contributions, in particular, often taking place at the same time
as income taxes, is likely to reinforce the regressive
impact of tax evasion.
Our approach relies on matching data from tax
returns with survey data. While we have made an
effort to make the two sources comparable, our
adjustment techniques offer at best good approximations. In particular, the truncated nature of tax
records (i.e. low-income families pay no taxes) and the
limited reliability of income statistics at either end of
the income scale leave our estimates vulnerable to
measurement error. Therefore, our results should be
best seen as tentative estimates under an experimental
research design.
Conclusion
The paper shows that tax evasion in Greece increases
inequality and poverty, and reduces tax progressivity,
while causing a considerable loss of tax receipts. Can
these findings be trusted? One cause for caution is the
distinction between the static and dynamic effects of
tax evasion. It is important to remember that taxation
(and, by implication, tax evasion) does not simply
reduce disposable incomes; it also affects decisions
concerning supply of, and demand, for labour, the
allocation of disposable income between consumption
and savings, the allocation of consumption between
different goods and services and so on (Slemrod and
Our key assumption is to treat incomes observed in
EU-SILC as closer approximations of ‘true income’
on the grounds that people have no incentive to conceal their income from survey interviewers, since their
disposable income would not be affected by their
response. The intuition is reasonable, but not neces-
31
CESifo Forum 2/2012
Focus
sarily correct. There are reasons to suspect that the
actual but unknown level of tax evasion may be considerably higher than that implied by our estimates. In
particular, there is some evidence that the very same
factors causing tax evasion, combined with the wish
of tax-evading individuals to be somehow ‘consistent’, may cause under-reporting in income surveys as
well, albeit at a lower level (Elffers et al. 1987).
Matsaganis, M. and C. Leventi (2011) The Distributional Impact of
the Crisis in Greece, Working Paper EM 3/11, Microsimulation Unit,
University of Essex.
OECD (2011), Economic Surveys: Greece, Paris.
Pashardes, P. and A. Polycarpou (2008), “Income Tax Evasion,
Inequality and Poverty”, Cyprus Economic Policy Review 2(2), 37–49.
Pissarides, C.A. and G. Weber (1989), “An Expenditure-based Estimate
of Britain’s Black Economy”, Journal of Public Economics 39, 17–32.
Sandmo, A. (2005), “The Theory of Tax Evasion: A Retrospective
View”, National Tax Journal 58, 643–663.
Schneider, F. (2012), The Shadow Economy and Work in the Shadow:
What Do We (Not) Know?, IZA Discussion Paper 6423.
In spite of the above caveats, we believe our results
capture essential aspects of the problem we set out to
explore. Our core finding, that tax evasion in Greece
has a regressive impact, seems reasonably robust. In
view of that, and under conditions of severe crisis, the
task of combating tax evasion assumes greater
urgency than ever before.
Slemrod, J. (2007), “Cheating Ourselves: The Economics of Tax
Evasion”, Journal of Economic Perspectives 21, 25–48.
Slemrod, J. and S. Yitzhaki (2002), “Tax Avoidance, Evasion, and
Administration”, in: Auerbach, A.J. and M. Feldstein (eds.),
Handbook of Public Economics, Amsterdam: Elsevier.
References
Andreoni, J., B. Erard and J. Feinstein (1998), “Tax Compliance”,
Journal of Economic Literature 36, 818–860.
Bank of Greece (2012), Governor’s Report for the Year 2011, Athens:
Bank of Greece.
Benedek, D. and O. Lelkes (2011), “The Distributional Implications
of Income Under-reporting in Hungary”, Fiscal Studies 32, 539–560.
CEU (2011a), Statement by the Heads of State or Government of the
Euro Area and EU Institutions (21 July 2011), Brussels.
CEU (2011b), Main Results of Euro Summit (26 October 2011),
Brussels.
Christian, C.W. (1994), “Voluntary Compliance with the Individual
Income Tax: Results from the 1988 TCMP Study”, IRS Research
Bulletin 1993/1994, Publication 1500, 35–42.
EC (2010), The Economic Adjustment Programme for Greece (first
review, summer 2010), European Economy Occasional Papers 61,
Luxembourg: Office for Official Publications of the European
Communities.
EC (2011), The Economic Adjustment Programme for Greece (third
review, winter 2011), European Economy Occasional Papers 77,
Luxembourg: Office for Official Publications of the European
Communities.
Elffers, H., R.H. Weigel and J. Dick (1987), “The Consequences of
Different Strategies for Measuring Tax Evasion Behaviour”, Journal
of Economic Psychology 8, 311–337.
Feldman, N. and J. Slemrod (2007), “Estimating Tax Noncompliance
with Evidence from Unaudited Tax Returns”, Economic Journal 117,
327–352.
Fiorio, C.V. and F. D’Amuri (2005), “Workers’ Tax Evasion in Italy”,
Giornale degli Economisti e Annali di Economia 64, 241–264.
IMF (2010), Greece: Staff Report on Request for Stand-by
Arrangement, Country Report 10/110, Washington DC.
IMF (2012), Greece: Staff Report on Request for Extended
Arrangement under the Extended Fund Facility, Country Report 12/57,
Washington DC.
Lyssiotou, P., P. Pashardes and T. Stengos (2004), “Estimates of the
Black Economy Based on Consumer Demand Approaches”,
Economic Journal 114, 622–640.
Marino, M.R. and R. Zizza (2008), L’evasione dell’Irpef: una stima
per tipologia di contribuente, Rome: Bank of Italy.
Matsaganis, M. and M. Flevotomou (2010) “Distributional
Implications of Tax Evasion in Greece”, Hellenic Observatory Papers
on Greece and Southeast Europe GreeSE Paper 31, The Hellenic
Observatory, LSE, London.
CESifo Forum 2/2012
32